Received June 11, 2016
Accepted September 23, 2016

Objective: To define the appropriate number of embryos to be transferred at day 5. Methods: Retrospective analysis of 784 consecutive fresh day-5 embryo transfers performed between 2007 and 2015, divided in three groups: Group A (N = 219): received the only 2 embryos that reached a transferable stage; Group B (N = 357): received 2 selected embryos among several that reached a transferable stage; Group C (N = 208): received the only 3 developing embryos. Clinical pregnancy, implantation, multiple pregnancy and delivery rates were registered. Kruskal-Wallis and Fisher Exact tests were applied as appropriate. Results: Age and previous attempts were comparable in the 3 groups. Compared with Group A, Groups B and C had a higher oocyte recovery (10.7 ± 5.6 vs. 14.7 ± 8.0 vs. 13.8 ± 6.6), fertilization rate (75.97% vs. 81.60% vs. 83.29%) and percentage of embryos reaching a transferable stage on day 5 (39.98% vs. 63.99% vs. 60.97%), as well as a significantly higher clinical pregnancy (42.92% vs. 61.06% vs. 58.17%) and implantation rates (21.09% vs. 40.98% vs. 36.97%). The multiple pregnancy rate was higher in Groups B and C than in Group A (11.70% vs. 31.19% vs. 37.19%). The high order multiple pregnancy rate (> 2) was significantly increased in group C (1.06% vs. 0.92% vs. 14.05%). Conclusions: In patients with 3 or more day 5 developing embryos, delivery rates are similar if 2 or 3 embryos are transferred. The transfer of 3 embryos carries an unacceptable increase in the risk of high order multiple pregnancy, with its known consequences. According to our data, we should not exceed the number of 2 day-5 fresh embryos transferred.